Because random assignment is not possible in observational studies, estimates of treatment effects might be biased due to selection on observable and unobservable variables. To strengthen causal inference in longitudinal observational studies of multiple treatments, we present 4 latent growth models for propensity score matched groups, and…

The purpose of this study was to evaluate the invariance of predictive relations across early-adolescent sex and ethnic groups regarding parenting factors and externalizing and internalizing problems and victimization. Data (n = 598; 54% female) from a triethnic (Hispanic, non-Hispanic white, and non-Hispanic black) probability sample of fifth…